Inevitably, people start getting impatient and hitting the door-close button after about 4 seconds. It doesn’t do anything, but the doors close a second or two later regardless, so people think they’ve affected the outcome, and they push the door-close button again the next time. If they push the button too soon, and the elevator waits a few more seconds before closing the doors, the people assume that it’s just being slow today or they didn’t hit the button hard enough.
They never consider the possibility that their action is not related to the result.
This is why superstition works. Animals learn it, too. “If I perform this action, I get this result.” It takes a more advanced or analytical mind to consider performing a test: “If I take no action, will I get this result anyway?”
I secretly think less of door-close people in the elevator.
Door close buttons in most elevators haven’t worked since around the early nineties. They’re there to trick passengers. There’s a lot of psychology involved. People tend to feel closed-in and helpless in elevators, which leads to claustrophobia and anxiety. Even something as small as allowing the passenger to think they have control over the doors can put them at ease.
There’s a surprising amount of deception involved in elevator design, actually, reinforcing the paseenger’s false belief that they have control, putting them at ease. They are, after all, in a box suspended by ropes. It’s all rather interesting. I recommend reading up on it. I can’t remember how I know even this much. There was a fascinating article on elevators in the New Yorker, I think, that the right Google queries might turn up.